In brief

Linus Torvalds says AI is becoming a useful tool for Linux kernel development, but not a replacement for programmers. Speaking at the Linux Foundation’s Open Source Summit North America, he said AI coding tools have helped drive a noticeable rise in kernel contributions, with recent releases seeing about 20% more commits than usual. He compared AI to earlier productivity tools such as assemblers and compilers, arguing that it can help generate code and find bugs, but developers still need to understand what the tools produce if they want to maintain serious software over time.

At the same time, Torvalds warned that AI is creating new pressure on open-source communities, especially around security reporting. The Linux kernel security mailing list has been flooded with duplicate or poorly handled AI-generated bug reports, forcing maintainers to spend time triaging issues rather than fixing them. Torvalds said AI-discovered vulnerabilities should often be treated as effectively public, because many others may find the same flaw, but he also urged researchers not to publish working exploits irresponsibly. His broader message is that AI can improve software quality and productivity, but it also creates social, operational, and security challenges that still require human judgment, coordination, and responsibility.

Source: ZDNET

‍

‍